系統(tǒng)生物學I-概述學習資料

Curriculumarrangement1.Introductiontosystemsbiology2.Advancedmeasurementplatformfor-omics3.Modeling&simulating4.CellularSimulation5.System-levelanalysis6.Systemsbiologysoftware1Lecture1：IntroductiontoSystemsBiology2OutlineIntroduction介紹2.MeasurementTechnology&ExperimentalMethods

測量技術和實驗方法3.Next-generationExperimentalSystems

下一代實驗系統(tǒng)4.RobustnessofBiosystems穩(wěn)健性5.Systeome系統(tǒng)體制6.ImpactofSystemsBiology系統(tǒng)生物學的影響7.Conclusion總結31.1.Ultimategoalofbiology？終極目標tounderstandeverydetailandprincipleofbiologicalsystems…；理解生物系統(tǒng)的每個細節(jié)和原則AncientfarmingCopyright1996byInterCityOz,Inc./magazine/mag07012001/magf5.htm4Mendel(1822-1884)GregorMendelpublishedhisfindingsonheredityinpeasinVersucheüberPflanzenHybriden.51900:

Mendel’sworkwasdiscovered6Morgan(1866-1945)1915:TheMechanismofMendelianHeredityChromosomesarethevehiclesofheredity.染色體是遺傳的載體Genesareonchromosomes.基因存在于染色體上Genescanchange(mutate).基因可變Whataregenesmadeof?A.H.SturtevantC.B.BridgesH.J.Muller7Schr?dinger(1887-1961)1944:

Schr?dingerasks“whatislife?”“……livingmatter,whilenoteludingthe‘lawsofphysics’asestablisheduptodate,

islikelytoinvolve‘otherlawsofphysics’

hithertounknown…...”

生活的物質，不逃避自古以來建立的物理定律，以及迄今為止沒有闡明的物理定律81944:DNAisthehereditarychemical!9Molecularstructureofnucleicacids:astructurefordeoxyribosenucleicacid.Nature171(1953):737–738.WatsonandCrick10MolecularBiologyMauriceWilkins1916-2004RosalindFranklin1920-1958FrancisCrick1916-2004JimWatson1928-1962:

NobelPrize111950-1970:

UnravellingofhowthecelldecodesDNA解開DNA細胞如何解碼DNA的機制---中心法則TheCentralDogmaofMolecularBiology121977:

determiningthesequenceofnucleotidesinDNA

確定DNA的核苷酸序列SangerGilbert1980:

FredSanger&WallyGilbertwereawardedtheNobelPrizeinChemistry.CTGGAAGAGGTATGTGCGCCGTTTCTGTTATCACAGTGTGCAATCCCATTACCGCATATCAGTTATAACAATAGTAATGGTAGCGCCATTAAAAATATTGTCGGTTCTGCAACTATCGCCCAATACCCTACTCTTCCGGAGGAAAATGTCAACAATATCAGTGTTAAATATGTTTCTCCTGGCTCAGTAGGGCCTTCACCTGTGCCATTGAAATCAGGAGCAAGTTTCAGTGATCTAGTCAAGCTGTTATCTAACCGTCCACCCTCTCGTAACTCTCCAGTGACAATACCAAGAAGCACACCTTCGCATCGCTCAGTCACGCCTTTTCTAGGGCAACAGCAACAGCTGCAATCATTAGTGCCACTGACCCCGTCTGCTTTGTTTGGTGGCGCCAATTTTAATCAAAGTGGGAATATTGCTGATAGCTCATTGTCCTTCACTTTCACTAACAGTAGCAACGGTCCGAACCTCATAACAACTCAAACAAATTCTCAAGCGCTTTCACAACCAATTGCCTCCTCTAACGTTCATGATAACTTCATGAATAATGAAATCACGGCTAGTAAAATTGATGATGGTAATAATTCAAAACCACTGTCACCTGGTTGGACGGACCAAACTGCGTATAACGCGTTTGGAATCACTACAGGGATGTTTAATACCACTACAATGGATGATGTATATAACTATCTATTCGATGATGAAGATACCCCACCAAACCCAAAAAAAATATGTGCAAAAAAATGCTTGATGATTTGTAATGAGATTGAGGAGGTTTCGAGACAGGCACCAAAGTTTTTACAAATGGAT13Molecularbiologyenablesustounderstandbiologicalsystemsasmolecularmachines,tounderstandtheelementaryprocessesbehindheredity,evolution,development,anddisease,includingreplication,transcription,translation,andsoforth.142001:TheHumanGenome分子生物學使我們了解生物系統(tǒng)作為生物機器的工作機制，了解遺傳、進化、發(fā)育、疾病背后的基本過程，包括復制、轉錄、翻譯等等。DNA:genome(mycoplasma,E.coli,C.elgans,Drosophila,Homosapiens)

基因組RNA:transcriptome&genesexpressionprofiling

轉錄組和基因表達譜Protein:proteome(bloodplasma,liver,brain…)蛋白質組Metabolite（代謝物）：metabolome代謝組Ourunderstandingtowardsmolecular-levelmechanismofbiologicalsystemsisaccelerating.Nevertheless,suchknowledgedosenotprovideuswithanunderstandingofbiologicalsystemsassystems.Genesandproteinsarethecomponentsofthesystems.Whatconstitutesthesystemisnecessaryforunderstandingthesystems,butnotsufficient.

我們對生物系統(tǒng)分子水平的了解日益加速，然而這些知識不能為我們提供生物作為一個系統(tǒng)的了解。基因和蛋白質是系統(tǒng)的組件，是我們了解系統(tǒng)構成的必要成分，但遠遠不夠151.2.Goalofsystemsbiology？

aimstodevelopasystem-levelunderstandingofbiologicalsystems.為了在系統(tǒng)水平上了解生物系統(tǒng)

Systemlevelunderstandingrequiresasetofprinciplesandmethodologies方法論thatlinksthebehaviorsofmoleculestosystemcharacteristicsandfunctions.Ultimately,cells,organisms,andhumanbeingswillbedescribedandunderstoodatthesystemlevelgroundedon基于aconsistentframework一致的框架ofknowledgethatisunderpinned支持、基礎bythebasicprinciplesofphysics.161.3.OriginofSystemsBiology1933：CannonConceptofHomeostasis內穩(wěn)態(tài)概念

Cannon,W.B.,(1933).Thewisdomofthebody,Norton,NewYork.

1948：WienerBiologicalCybernetics生物控制論

Wiener,N.,(1948).CyberneticsorControlandCommunicationintheAnimalandtheMachine,TheMITPress,Cambridge.1968：vonBertalanffyGeneralSystemsTheory一般系統(tǒng)理論

vonBertalanffy,L.(1968).GeneralSystemTheory,Braziler,NewYork.Withthelimitedavailabilityofknowledgefrommolecularbiology,mostsuchattemptshavefocusedonthedescriptionandanalysisofbiologicalsystemsatthephysiologicallevel.生理水平1718ModernSystemsBiology現代系統(tǒng)生物學Modernsystemsbiologythatdistinguishesitfrompastattemptsisthatthereareopportunitiestoground接觸，基礎system-levelunderstandingdirectlyonthemolecularlevelsuchasgenesandproteins,whereas然而pastattemptshavenotbeenabletosufficientlyconnectsystemleveldescriptiontomolecular-levelknowledge.

Thus,althoughitisnotthefirsttimethatsystem-levelunderstandinghasbeenpursued繼續(xù)從事,itisthefirsttimetohaveanopportunitytounderstandbiologicalsystemswithintheconsistentframework一致框架ofknowledgebuiltupfromthemolecularleveltothesystemlevel.19CharacteristicsofModernSystemsBiologyComprehensiveanalysisofallthemoleculesinsidethecell對細胞里全部分子的綜合分析Measurementofdynamicchangesofallthemolecules對所有分子進行動態(tài)測量Dataintegrationwithmathematicorcomputationalmethod用數學或計算的方法進行數據整合Dynamicanalysisofbiosystemsfrommolecularleveltoorganism對生物系統(tǒng)從分子水平到有機體水平的動態(tài)分析Modelingbasedonsystemscience系統(tǒng)科學基礎上建模Combinationofdiscoveryscienceandhypothesis-drivenscience對已發(fā)現的科學現象和假設的現象結合201.4.Definition定義HiroakiKitano

LeroyHood

Systemsbiologyrepresentsanapproachtoinvestigate調查allthecomponents(genes,mRNAs,proteins,cells,tissuesandorgansetc.)ofabiologicalsystem,andtheirinteractionsinspecificconditions,thusto由此realizetheintegrationofinformationofdifferentlevelsfromgenestocells,tissuesandorganism.

Integration&Quantitation定量21

BiologicalSystems22生命組學，生理組學ComponentInteraction1022313系統(tǒng)復雜性

Forasystem,themostimportantthingisnotthefunctionsoftheindividuals,buttheirinteractions！23Methodtoinvestigatecomplexsystem探究復雜系統(tǒng)的方法1.Reductionism還原論Ontologicalreductionism本體論還原Methodologicalreductionism

方法論還原2.Holism整體論3.Generalsystemstheory一般體系論

Chaostheory混沌理論Cellularautomata細胞自動控制器Catastrophetheory突變理論Hierarchicalsystems分級系統(tǒng)

24ChaostheoryPredictability:DosetheflapofaButterflyWingsinBrazilsetoffaTornadoinTexas?(1972)可預言性EdwardLorenzWhenthepresentdeterminesthefuture,buttheapproximate大概的presentdoesnotapproximatelydeterminethefuture.Unpredictability不可預測性Extremelysensitivetotheinitialcondition對最初的條件十分敏感25SYamanakaNature

460,49-52(2009)StochasticmodelApplication:stochasticmodelingforinducedpluripotentstemcellgeneration應用：誘導多功能干細胞的隨機模型26Cellularautomata細胞自動機Acellularautomataconsistsofaregulargrid規(guī)則網格ofcells,eachinoneofafinitenumber有限數ofstates,suchasonandoff.Thegridcanbeinanyfinitenumberofdimensions規(guī)模.Foreachcell,asetofcellscalleditsneighborhoodisdefinedrelativetothespecifiedcell.Aninitialstate(timet=0)isselectedbyassigningastateforeachcell.Anewgenerationiscreated(advancingtby1),accordingtosomefixedrule(generally,amathematicalfunction)thatdeterminesthenewstateofeachcellintermsofthecurrentstateofthecellandthestatesofthecellsinitsneighborhood.Typically,theruleforupdatingthestateofcellsisthesameforeachcellanddoesnotchangeovertime,andisappliedtothewholegridsimultaneously.JohnvonNeumannCharacteristics:parallelcomputation,local,homogeneousStephenWolframcurrentpatternnewstatusofcentercell27ApplicationProtNet:atoolforstochasticsimulationsofproteininteractionnetworksdynamics一種對蛋白質相互作用動態(tài)網絡進行模擬的工具BMCBioinformatics.2007;8(Suppl1):S4.28Catastrophetheory突變理論Catastrophetheory,whichoriginatedwiththeworkoftheFrenchmathematicianRenéThominthe1960s,andbecameverypopularduetotheeffortsofChristopherZeemaninthe1970s,considersthespecialcasewherethelong-runstableequilibriumcanbeidentifiedwiththeminimumofasmooth,well-definedLyapunovfunction.Smallchangesincertainparametersofanonlinearsystemcancauseequilibriatoappearordisappear,ortochangefromattractingtorepellingandviceversa,leadingtolargeandsuddenchangesofthebehaviourofthesystem.RenéThom

FieldsMedalin1958ChristopherZeemanWhichtheoryismoreaccurate?Better?29Application:MassivegenomicrearrangementacquiredinaSingleCatastrophicEventduringCancerDevelopment在癌癥發(fā)展過程中單一災難性事件中收集到的大量基因組重排FISHProfilingofTK10GenomicFeaturesofChromothripsisSuggestthatMostRearrangementsOccurinaSingleCatastrophicEventPhilipJ.Stephens,etal.Cell144,27–40,201130Hierarchicalsystems分級系統(tǒng)1)Itiseasiertoanalyseand(re)designlarge-scalesystemsiftheyarebrokenintosmallerunits.2)Thesub-systemapproachallowsspecialisation,whereasub-systemisonlyresponsibleforitsowntaskanddoesnotrequireinformationoftheobjectivesoftheoverallsystem.3)Hierarchicalsystemsallowacertaindegreeoffaulttolerance.Thisisduetothefactthatifasub-systembreaksdown,theoverallsystemdoesnotnecessarilystopworking.Furthermore,dueto’modulestructure’thefailureis’localised’andhenceeasytodetectandandrepair.Thecoordinator,however,istheweakpoint,becauseifitstopsworking,theoverallsystemcannotfunctionanymore.4)Evenevolutionseemstofavourtwo-levelhierachicalsystems.Forexampleinahumanbodythebraincanbeconsideredasbeingthecoordinator,whereastherestofthebodyformsthesub-systemlevel.5)Intheevolutionoforganisationstwo-levelhierachicalsystemsplayamajorrole.Evenpre-historictribeshadatribeleader,whomwasresponsibleforcoordinatingtheactionsofindividualtribemembersinordertoimprovetheoverallwell-beingofthetribe.Whytwolevelhierarchicalsystemsaresofrequent?Complexity&Complicatedness復雜度和復雜性31Application:OscillationsinNF-kBSignalingControltheDynamicsofGeneExpressionNF-kBSignaling的震動控制基因表達動力學D.E.Nelsonetal.Science306,704(2004)32SimplestmodelAveragecomplexitymodelMostcomplexmodel33PropertyofBiosystem系統(tǒng)生物的性質Emergence出現Anyfunctionperformedbyasystemthatisnottheresultofasinglepartinthesystem,butratheristheresultofinteractingpartsinthesystem.1)Brandnewproperty:全新的屬性Theinteractionofthesepartsgivesrisetonewpropertiesandfunctions.2)Irreducible:

不能削減Systemscannotbereducedto分解為theirindividualpartsorstudiedonepartatatime.3)Unpredictable:不可預測Systems’characteristicscannotbepredictedfromtheirindividualparts.Robustness穩(wěn)健性Thesystemmaintainsitsfunctionalpropertiesagainstvariousdisturbances.干擾1)Adaptation;適應性

2)Parameterinsensitivity;參數不敏感

3)Gracefuldegradation.功能降低34Scale-free無標度Ascale-freenetwork’s無標度網絡degreedistributionfollowsapowerlaw冪次定律,thatis,thefractionP(k)ofnodesinthenetworkhavingkconnectionstoothernodesgoesforlargevaluesofkas:p(k)~k-n

wherenisaparameterwhosevalueistypicallyintherange2<n<3.

InternetProtein-proteinnetwork35PhysicsChemistryBiologyTraditionalMethodology:Reductionism還原論，簡化倫36Philosophyshiftsfromreductionistictointegrative37Science2002,763-764Systemsbiology:acombinationofdiscoveryscience&hypothesis-drivenscience381.Systemstructureidentification2.Systembehavioranalysis3.Systemcontrol4.Systemdesign1.5.Mission任務391.5.1.Identificationofsystemstructure系統(tǒng)結構識別1.Genome,transcriptome,proteome,metabolome…;2.Regulatoryrelationshipsamonggenes;3.Interactionsofproteinsthatprovidesignaltransductionandmetabolismpathways;4.Physicalstructuresoforganisms,cells,organelle,chromatin,andothercomponents;5.Physicalstructuresofwholeanimalsprecisely精確地atthecellularlevelformulticellularorganism.401.5.2.Systembehavioranalysis

Onceasystemstructureisidentifiedtoacertaindegree,itsbehaviorneedstobeunderstood.e.g.Sensitivityofcertainbehaviorsagainstexternalperturbations干擾;Howquicklythesystemreturnstoitsnormalstateafterthestimuli刺激.Significance:意義Reveal揭露system-levelcharacteristics;Provideimportantinsights見解formedicaltreatmentsbydiscoveringcellresponsetocertainchemicals.411.5.3.Systemcontrol系統(tǒng)控制

Byapplyingtheinsightsobtained獲得bysystemstructureandbehaviorunderstanding,wecanestablishamethodtocontrolthestateofbiologicalsystems.e.g.Transformmalfunctioning出故障的cellsintohealthycells;Controlcancercellstonormalcellsorcauseapoptosis細胞凋亡;Controlthedifferentiation分化status狀態(tài)ofaspecificcellintoastemcell干細胞,andcontrolittodifferentiateintothedesiredcelltype.SignificanceTechnologiestoaccomplishsuchcontrolwouldenormouslybenefithumanhealth.421.5.4.Systemdesign

Ultimately,wewouldliketoestablishtechnologiesthatallowustodesignbiologicalsystemswiththeaimofprovidingcuresfordiseases.e.g.Designandgroworgansfromthepatient’sowntissue.Suchanorgancloningtechniquewouldbeenormouslyusefulforthetreatmentofdiseasesthatrequireorgantransplants.器官移植Engineeringapplicationsbyusingbiologicalmaterialsforroboticsorcomputation.Byusingmaterialsthathaveself-repairandself-sustainingcapability自我維持能力,industrialsystemswillberevolutionized.43441.6.Keyfieldsinsystemsbiology關鍵領域1.Biologicalscience:molecularbiology&-omics組學2.Computationscience：

simulation、bioinformatics&software3.Systemscience:systemanalysisbasedoncontroltheory4.Analyticalscience:highthroughput,precisemeasurement4546472.1.ComprehensivemeasurementGenomesequence:capillaryarrayelectrophoresis毛細血管微陣列電泳Geneexpressionprofiling:

biochip,RTPCR基因表達譜Proteinsequence:separation&MS蛋白質序列Interaction:twohybrid,TAP,biochipProteinlocalization:molecularimaging蛋白質定位Metabolite代謝物：separation&MS,NMRModelspecies:Yeast,Celegan,Drosophila

482.2.MeasurementforSystemsBiologyHigh-throughput高通量,comprehensive綜合的,andaccuratemeasurement精確測量isthemostessentialpartofSB.Itwillnevergenerateseriousresultswithoutexperimentaldatauponwhichcomputationalstudiescanbegrounded.有充足的理由的Tobesuccessful,measurementhastobe(1)comprehensive,綜合性(2)quantitativelyaccurate,定量準確(3)systematic.系統(tǒng)的492.2.1.Comprehensivemeasurement？

1.factorcomprehensivenessComprehensivenessintermsof依據targetfactorsthatarebeingmeasured,suchasnumbersofgenesandproteins.Itisimportantthatmeasurementiscarriedoutintensively強烈地，集中地forthefactorsthatarerelatedtothecentralgenesandproteinsofinterest.Unlessallgenesandproteinsaremeasured,howeffectivelymeasurementcoversthefactorsofinterestismoreimportant,ratherthanthesheer絕對的numberoffactorsmeasured.e.g.toinfer推斷geneticregulatorynetworksfromanexpressionprofile概況502.time-series時間序列comprehensivenessInmodelingandanalysisofadynamicalsystem,itisimportanttocaptureitsbehaviorwithfine-grain細粒timeseries.Traditionalbiologicalexperimentstendtoensureonlythechangebeforeandafteracertainevent.Forcomputationalanalysis,datameasuredataconstanttimeintervalareessential基本的inadditiontotraditionalsamplingpoints.樣本點e.g.cellularaging細胞衰老513.itemcomprehensivenessTherearecaseswhereseveralfeatures,suchastranscriptionlevel,proteininteraction,phosphorylation,localization,andotherfeatures,havetobemeasuredintensivelyforthespecifictarget.e.g.proteininteractions&expressionprofiles522.2.2.Quantitativelyaccuratemeasurement？Tounderstanddynamics,itisessentialthateachparameterregardingthenetworkisobtained,sothatvariousnumericalsimulationsandanalysescanbeperformed.Suchparametersarebindingconstant,transcriptionrate,translationrate,chemicalreactionrate,degradationrate,diffusionrate擴散率,speedofactivetransport主動轉運,etc.Whileaccuracyisimportant,thelevelofaccuracyrequiredmayvarydependingonwhichpartofthesystemistobemeasured.53“accuracy”and“precision精度”542.2.3.Systematicmeasurement？“Systematic”meansthatmeasurementisperformed執(zhí)行insuchawaythatobtained獲得的datacanbeconsistentlyintegrated.Theidealsystematicmeasurementissimultaneousmeasurementofmultiplefeaturesforasinglesample單樣本.

Although

thisrequirementsoundsobvious,veryfewdatasetsmeetthesecriteria

標準today.55Highthroughput高通量Automatic自動化Highspeed高速Highaccurate高準確性Lowcost低消耗最大限度提高單位時間內消耗單位資源所獲取的生物信息數量與質量。56Next-generationExperimentalSystemse.g.InthedevelopmentalbiologyofC.elegans線蟲,identificationofcelllineage細胞系isoneofthemajorissuesthatneedstobeaccomplishedtoassistanalysisofthegeneregulatorynetworkfordifferentiation.Thefirstattempttoidentifycelllineagewascarriedoutentirelymanually手動的.ittookseveralyearstoidentifythelineageofthewildtype.Sulston,J.E.,Schierenberg,E.,White,J.G.,andThomson,J.N.(1983).TheembryoniccelllineageofthenematodeC.elegans.Dev.Biol.100:64–119.Sulston,J.E.andHorvitz,H.R.(1997).Post-embryoniccelllineageofthenematodeC.elegans.Dev.Biol.56:110–156.57C.Elegan線蟲s---asmartanimalBrennerHorvitzSulston2002NobelPrizeFireMello2006NobelPrizeSimple,Open,Clever&Correlative---modelanimalforgenetics,neuroscience&developmentalbiology1998Wholegenomesequence58Onami,S.,et.al.(2001).Automaticacquisitionofcelllineagethrough4DmicroscopyandanalysisofearlyC.elegansembryogenesis.FoundationsofSystemsBiology,TheMITPress,Cambridge.Withtheavailabilityofexhaustive徹底的RNAiknockoutforC.elegans,high-throughputcelllineageidentificationisessentialtoexploretheutility效用oftheexhaustiveRNAi核苷酸干擾.Effortsareunderwaytofullyautomatecelllineageidentification.aswellasthreedimensional三維的nuclei核心positiondataacquisition數據采集,fullyUtilizing利用advancedimageprocessingalgorithms處理算法andmassivelyparallelsupercomputers大規(guī)模并行超級計算機.593.1.Next-generationexperimentaltechnologyMicroarraychip微陣列芯片Microfluidicchip微流體芯片Nanotechnology納米技術Femtosecondtechnology飛秒技術……60Human22KOligo-chipscreeningbreastcancerbiomarkergenesI.Microarraychip

Molecularlevel:genechipproteinchip61I.Microarraychip

Cellularlevel:cellchip62I.Microarraychip

Tissuelevel:tissuechip63II.Microfluidicchip19462002200364III.Biomassspectrometer生物質譜儀65IV.Molecularimaging663.2.SystemstructureidentificationMajortasks：

I.MolecularelementII.NetworkstructureIII.ParameterMethodology：

I.bottom-upII.top-down673.2.1.Bottom-up自底而上Toconstructageneregulatorynetworkbasedonthecompilation編輯ofindependentexperimentaldata,mostlythroughliteraturesearches文獻檢索andsomespecificexperimentstoobtaindataofveryspecificaspectsofthenetworkofinterest.lambdaphagedecisioncircuit判定電路earlyembryogenesis早期胚胎形成ofDrosophilaMcAdams,H.andShapiro,L.(1995).CircuitSimulationofgeneticnetworks.Science269:650–656.Reinitz,J.,Mjolsness,E.,andSharp,D.H.(1995).ModelforcooperativecontrolofpositionalinformationinDrosophilabybicoidandmaternalhunchback.J.Exp.Zoo.271:47–56.68Phagelamdalysis–lysogenydecisioncircuit69Applicabilityofbottom-upapproach適用性Thisapproachissuitablewhenmostofthegenesandtheirregulatoryrelationshiparerelativelywellunderstood.Thisapproachisparticularlysuitablefortheend-gamescenario終極場景wheremostofthepiecesareknownandoneistryingtofindthelastfewpieces.703.2.2.Top-down自上而下Thetop-downapproachtriestomakeuseofhigh-throughputdatausingDNAmicroarrayandothernewmeasurementtechnologies.theyeastcellcycledevelopmentofmousecentralneuralsystems中樞神經系統(tǒng)Brown,P.O.andBotstein,D.(1999).ExploringthenewworldofthegenomewithDNAmicroarrays.NatureGenetics21:33–37.D’haeseleer,et.al(1999).LinearmodelingofmRNAexpressionlevelsduringCNSdevelopmentandinjury.Proc.PacificSymposiumonBiocomputing’99pp.41–52.713.2.3.ParameterIdentificationItisimportanttoidentifyonlythestructureofthenetwork,butasetofparameters,becauseallcomputationalresultshavetobematchedandtestedagainstactual真實的experimentalresults.Inmostcases,theparametersethastobeestimatedbasedonexperimentaldata.Variousparameteroptimization參數優(yōu)化methods,suchasbruteforceexhaustivesearch窮舉搜索,geneticalgorithms基因演算法andsimulatedannealing模擬退火,areusedtofindasetofparametersthatcangeneratesimulationresultsconsistentwithexperimentaldata.723.3.SystembehavioranalysisHowdoesitadapttochangesintheenvironment,suchasnutrition,andvariousstimuli?Howdoesitmaintainrobustnessagainstvariouspotentialdamagetothesystem,suchasDNAdamageandmutation?Howdospecificcircuitsexhibitfunctionsobserved?Toattainsystem-levelunderstanding,itisessentialtounderstandthemechanismsbehind(1)therobustnessandstabilityofthesystem；(2)functionalitiesofthecircuits.733.4.SimulationSimulationofthebehaviorofgeneandmetabolismnetworksplaysanimportantroleinsystemsbiologyresearch;Simulationisanessentialtoolnotonlyforunderstandingthebehavior,butalsoforthedesignprocess.Forsimulationtobeaviablemethodologyforthestudyofbiologicalsystems,highlyfunctional,accurate,anduser-friendlysimulatorsystemsneedtobedeveloped.Thesimulatorneedstobecoupledwithparameteroptimizationtools,ahypothesisgenerator,andagroupofanalysistools.74Softwaresystemsforsystemsbiologyincludes:?databaseforstoringexperimentaldata?cellandtissuesimulator?parameteroptimizationsoftware?bifurcationandsystemsanalysissoftware?hypothesesgeneratorandexperimentplanningadvisorsoftware,and?datavisualizationsoftware.75SystemsBiologyWorkbench，SBWtoprovideacommonmiddlewaresothatplug-inmodulescanbeaddedtoformauniformsoftwareenvironment.SystemsBiologyMark-upLanguage,SBMLaversatileandcommonopenstandardthatenablestheexchangeofdataandmodelinginformationamongawidevarietyofsoftwaresystems.76SBW:旨在提供一個通用中間件用以整合多個插件模塊構成一個統(tǒng)一軟件環(huán)境。SBML:一個多用途且通用的開放標準，允許在大量的軟件系統(tǒng)之間交換數據和建模。它是XML的擴展，有希望成為數據和模型交換格式方面的工業(yè)與學術標準。3.5.AnalysismethodFBABABorisuk,M.andTyson,J.(1998).Bifurcationanalysisofamodelofmitoticcontrolinfrogeggs.JournalofTheoreticalBiology195:69–85.MCAFell,D.A.(1996).Understandingthecontrolofmetabolism,PortlandPress,London.Edward,J.S.andPalsson,B.O.(1999).SystemspropertiesoftheHaemophilusinfluenzaeRdmetabolicgenotype.JournalofBiologicalChemistry274:17410–17416.77

Robustnessisoneoftheessentialfeaturesofbiologicalsystems.Understandingthemechanismbehindrobustnessisparticularlyimportantbecauseitprovidesin-depthunderstandingonhowthesystemmaintainsitsfunctionalpropertiesagainstvariousdisturbances.透過它可以深入理解系統(tǒng)如何在各種擾動情況下維持自身功能特性：(1)changesinenvironment;(2)internalfailures78RobustnessofBiologicalsystems4.1.LessonsfromComplexEngineeringSystemsinterestinganalogiesbetweenbiologicalsystems&engineeringsystems:1.designedincrementallythroughsomesortofevolutionaryprocesses;2.generallysuboptimalforthegiventask;3.increasedcomplexitytoattainahigherlevelofrobustnessbandstability.e.g.AirplaneMycoplasma794.2.Howtorealizetherobustness?systemcontrolredundancy冗余modulardesign模塊化設計structuralstability結構穩(wěn)定性804.2.1.Control

Theuseofexplicitcontrolschemeisaneffectiveapproachtoimprovingrobustness.Feedforwardandfeedbackaretwomajormethodsofsystemcontrol.81e.g.1：E.colichemotaxisYi,T.-M.,et.al(2000).Robustperfectadaptationinbacterialchemotaxisthroughintegralfeedbackcontrol.PNAS97(9):4649–4653.82e.g.2：HeatshockcircuitHeatshockresponsewithfeedforwardandfeedbackcontrol83e.g.3：Growthcontrolofhumancellsp53relatedfeedbackloop844.2.2.Redundancy冗余在獲取系統(tǒng)魯棒性及處理系統(tǒng)組件意外損壞方面起重要作用。MAP激酶級聯反應

MAP激酶級聯反應包括并聯的通路之間廣泛的相互作用。即使因為變異或其他原因導致通路中的一條失效，MAP激酶途徑仍能通過其他通路轉導信號從而維持其功能。

Redundancyalsoplaysanimportantroleinattainingrobustnessofthesystem,andiscriticalforcopingwithaccidentaldamagetocomponents.RedundancyinMAPkinasecascade854.2.3.ModularDesignModulardesignisacriticalaspectoftherobustness:itensuresthatdamageinonepartofthesystemdoesnotspreadtotheentiresystem.Itmayalsoensureefficientreconfiguration重新配置throughouttheevolutionaryprocesstoacquirenewfeatures.Thecellularstructureofthemulticellularorganismisacleare